Uma Manikanta Vabilisetti
AI Engineer | Generative AI & Multi-Agent Systems | LLMs · RAG · MLOps | AWS & Azure | MPS Data Science @ UMBC
- Role
- Ai Engineer at Quadrant Technologies
- Location
- Redmond, WA, US
- LinkedIn followers
- 500 followers
About Uma Manikanta Vabilisetti
AI Engineer | Generative AI · Multi-Agent Systems · MLOps · AWS & Azure I build AI systems that actually work in production.With 4+ years of experience across Generative AI, machine learning, and cloud-native development, I specialize in turning complex AI research into scalable, enterprise-ready solutions. My work spans LLM integration, RAG pipeline architecture, multi-agent orchestration, computer vision, and MLOps — delivering measurable outcomes that go beyond demos and prototypes. Master\'s in Data Science — University of Maryland Baltimore County (GPA: 3.7/4.0) WHAT I BUILD Agentic AI Systems — Multi-agent pipelines (MAF, LangGraph, GPT-4.1) automating end-to-end enterprise workflows with deterministic business logic RAG Pipelines — Vector search, embedding generation, semantic reranking, and LLM validation agents for intelligent document processing and deduplication MLOps Platforms — End-to-end ML lifecycle on Azure ML SDK v2 with MLflow, GPU cluster orchestration, and automated artifact pipelines AI-Enabled Automation — Intelligent monitoring, auto-remediation, incident summarization, and runbook generation using Claude Code, GitHub Copilot & AWS Bedrock Computer Vision — YOLO-based object detection achieving 99.5% mAP50 on enterprise document classification RECENT HIGHLIGHTS Reduced manual logistics monitoring by ~95% with a 4-agent GPT-4.1 orchestration system 99.5% mAP50 detection accuracy — YOLO on Azure ML / A100 GPU Enterprise RAG + GPT-4o validation agent for duplicate detection (80%+ precision) 90%+ extraction accuracy on AI-powered document processing & compliance automation CORE STACKLLMs: GPT-4.1, GPT-4o, Claude, AWS Bedrock AgentsFrameworks: LangChain, LangGraph, MAF, FastAPI, StreamlitCloud: Azure (OpenAI, ML, Functions, App Service, AI Search, Azure Machine Learning, AI Serach, Key Valult, Container Registery, App Service, Storage Account) · AWS (Bedrock, S3, Lambda, EKS, ECR)MLOps: Azure ML SDK v2, MLflow, Docker, Azure Container RegistryAI Tools: Claude Code, GitHub CopilotLanguages: Python · SQL · BashOpen to connecting with AI practitioners, engineering leaders, and teams building intelligent systems. v••••••••@gmail.com
Experience
Ai Engineer
Sep 2024 — Present · Hyderabad, IN
Building and deploying production-grade Generative AI, multi-agent orchestration, and computer vision systems on Azure and AWS. PROJECT HIGHLIGHTS Multi-Agent Logistics Automation (AI-Enabled DevOps)Architected a 4-agent GPT-4.1 orchestration system (MAF GroupChat) that auto-detects supply chain delays from Microsoft Fabric Lakehouse (Gold Layer SQL), applies 60-min grace periods & ETA lookback logic, and dispatches threaded alerts via Microsoft Graph API — reducing manual monitoring by ~95%. Exposed as FastAPI REST API; containerized with Docker on Azure App Service. MLOps — Geotechnical Document AI (YOLO on Azure ML)End-to-end MLOps pipeline on Azure ML SDK v2: versioned Docker/conda environments, MLflow tracking, A100 GPU cluster management. Trained 3 YOLO variants achieving 99.5% mAP50 / 0.944 mAP50-95. Cut wasted compute ~35% via early stopping & batch scaling (8→4→2). RAG Pipeline — AI Duplicate DetectionSQL → Azure OpenAI embeddings → Azure AI Search (k-NN) → custom Python reranker → GPT-4o validation agent (MERGE/REVIEW, 80%+ precision). Self-serve FastAPI + Streamlit UI. Intelligent Document Processing & Compliance AutomationServerless pipeline (Azure Functions + Blob Storage) with Azure Document Intelligence + GPT compliance prompting — 90%+ JSON extraction accuracy on PDF/PNG documents. Grocery Flyer Extraction — GPT-4.1 VisionSpatial mapping algorithm for price-product association across multi-page PDFs (90%+ accuracy). Streamlit human-in-the-loop review + Azure AI Search with deduplication. GPT-4.1/4o · Claude Code · LangChain · LangGraph · MAF · Azure ML · MLflow · YOLO · Docker · FastAPI · Azure OpenAI · Microsoft Fabric · Python
Education
University of Maryland Baltimore County
Master of Science - MS
KL University
Bachelor of Science
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.